Jennifer Lindsey Suder Klettlinger Glenn Research Center, Cleveland, Ohio Fischer-Tropsch Cobalt Catalyst Improvements With the Presence of TiO 2 , La 2 O 3 , and ZrO 2 on an Alumina Support NASA/TM—2012-216020 October 2012 https://ntrs.nasa.gov/search.jsp?R=20130000071 2018-04-23T13:10:50+00:00Z
102
Embed
Fischer-Tropsch Cobalt Catalyst Improvements , La O, · PDF fileJennifer Lindsey Suder Klettlinger Glenn Research Center, Cleveland, Ohio Fischer-Tropsch Cobalt Catalyst Improvements
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Jennifer Lindsey Suder KlettlingerGlenn Research Center, Cleveland, Ohio
Fischer-Tropsch Cobalt Catalyst ImprovementsWith the Presence of TiO2, La2O3, and ZrO2 onan Alumina Support
Since its founding, NASA has been dedicated to theadvancement of aeronautics and space science. TheNASA Scienti c and Technical Information (STI)program plays a key part in helping NASA maintainthis important role.
The NASA STI Program operates under the auspicesof the Agency Chief Information Of cer. It collects,organizes, provides for archiving, and disseminatesNASA’s STI. The NASA STI program provides accessto the NASA Aeronautics and Space Database andits public interface, the NASA Technical ReportsServer, thus providing one of the largest collectionsof aeronautical and space science STI in the world.Results are published in both non-NASA channelsand by NASA in the NASA STI Report Series, whichincludes the following report types:
• TECHNICAL PUBLICATION. Reports ofcompleted research or a major signi cant phaseof research that present the results of NASAprograms and include extensive data or theoreticalanalysis. Includes compilations of signi cantscienti c and technical data and informationdeemed to be of continuing reference value.NASA counterpart of peer-reviewed formalprofessional papers but has less stringentlimitations on manuscript length and extent ofgraphic presentations.
• TECHNICAL MEMORANDUM. Scienti cand technical ndings that are preliminary orof specialized interest, e.g., quick releasereports, working papers, and bibliographies thatcontain minimal annotation. Does not containextensive analysis.
• CONTRACTOR REPORT. Scienti c andtechnical ndings by NASA-sponsoredcontractors and grantees.
• CONFERENCE PUBLICATION. Collectedpapers from scienti c and technicalconferences, symposia, seminars, or othermeetings sponsored or cosponsored by NASA.
• SPECIAL PUBLICATION. Scienti c,technical, or historical information fromNASA programs, projects, and missions, oftenconcerned with subjects having substantialpublic interest.
• TECHNICAL TRANSLATION. English-language translations of foreign scienti c andtechnical material pertinent to NASA’s mission.
Specialized services also include creating customthesauri, building customized databases, organizingand publishing research results.
For more information about the NASA STIprogram, see the following:
• Access the NASA STI program home page athttp://www.sti.nasa.gov
Jennifer Lindsey Suder KlettlingerGlenn Research Center, Cleveland, Ohio
Fischer-Tropsch Cobalt Catalyst ImprovementsWith the Presence of TiO2, La2O3, and ZrO2 onan Alumina Support
NASA/TM—2012-216020
October 2012
National Aeronautics andSpace Administration
Glenn Research CenterCleveland, Ohio 44135
Acknowledgments
I would like to express my gratitude to my advisor Dr. Steven S. C. Chuang for his guidance and support during my thesis research.I would also like to thank Dr. Bi-min Zhang Newby and Dr. George Chase who were my committee members. I appreciate theCenter for Applied Energy Research at The University of Kentucky, namely Dr. Burt Davis and Dr. Gary Jacobs for their guidanceand continued support in catalyst characterization. I would also like to express my appreciation to my colleagues at NASA GlennResearch Center: Chia (Judy) Yen, Angela Surgenor, Leah Nakley, Rachel Rich, Ana De La Ree, Lauren Best, and Al Hepp fortheir understanding and patience during my completion of this work. I am appreciative of the support from my supervisor, Dr. Chi-Ming Lee and the funding received from NASA Glenn Research Center for completion of my thesis and graduate coursework. I amdeeply grateful to my husband for his ongoing love and encouragement, as well as my parents for their dedication, encouragement,and many years of support throughout my life. The work in this thesis would not have been possible without those mentioned inthe above acknowledgment.
Available from
NASA Center for Aerospace Information7115 Standard DriveHanover, MD 21076–1320
National Technical Information Service5301 Shawnee Road
Alexandria, VA 22312
Available electronically at http://www.sti.nasa.gov
Trade names and trademarks are used in this report for identi cationonly. Their usage does not constitute an of cial endorsement,either expressed or implied, by the National Aeronautics and
Space Administration.
This work was sponsored by the Fundamental Aeronautics Programat the NASA Glenn Research Center.
Level of Review: This material has been technically reviewed by technical management.
Supplementary Notes
This report was submitted as a thesis in partial ful llment of the requirements for the degree of Master of Science to thegraduate faculty of the University of Akron, Akron, Ohio, May 2012.
3.2.1 Scanning Electron Microscopy and Electron DispersedSpectroscopy ....................................................................................................... 29
IV. CATALYST CHARACTERIZATION RESULTS ...................................................................... 34
4.1 Scanning Electron Microscopy and Electron DispersedSpectroscopy ....................................................................................................... 34
4.2 Brunauer Emmet Teller (BET)4 and Barrett Joyner Halenda (BJH)33
APPENDIX A. Pore Size Distribution Profiles ............................................................................ 80
NASA/TM—2012-216020 vii
LIST OF TABLES
Table Page
2.1 BET Data from Literature ........................................................................................................ 16
2.2 Hydrogen Chemisorption Data from Literature ............................................................. 20
2.3 XRD Data from Literature ........................................................................................................ 23
3.1 Catalyst support composition................................................................................................ 25
4.1 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)quantitative results for 9.7%TiO2-Al2O3 ............................................................................ 38
4.2 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)quantitative results for 3.0%La2O3-Al2O3.......................................................................... 41
4.3 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)quantitative results for 3.1%ZrO2-Al2O3. ........................................................................... 44
4.4 BET surface area measurements and BJH pore volume and pore radiusmeasurements.............................................................................................................................. 49
4.5 Hydrogen chemisorption / pulse reoxidation results following 10 hourhydrogen reduction at 350oC. ................................................................................................ 64
4.6 X-ray Diffraction results from XRD plots and Scherrer Equation ............................ 70
4.1 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 9.7%TiO2-Al2O3 where Al2O3 is pink (left) and TiO2 isgreen (right). ........................................................................................................................... 35
4.2 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results overlayed on scanning electron microscopy image for9.7%TiO2-Al2O3 where Al2O3 is pink (left) and TiO2 is green (right). ............... 36
4.3 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 9.7%TiO2-Al2O3 overlaid on scanning electronmicroscopy image. Alumina is highlighted in pink and titania is highlightedin green. ..................................................................................................................................... 37
4.4 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)quantitative results for 9.7%TiO2-Al2O3 support. .................................................... 38
4.5 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 3.0%La2O3-Al2O3 where Al2O3 is pink (left) and La2O3 isgreen (right).. .......................................................................................................................... 39
4.6 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results overlaid on scanning electron microscopy image for3.0%La2O3-Al2O3 where Al2O3 is pink (left) and La2O3 is green (right). .......... 40
4.7 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 3.0%La2O3-Al2O3 overlaid on scanning electronmicroscopy image (Al203 in pink and La2O3 in green). ........................................... 40
4.8 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)Quantitative Results for 3.0%La2O3-Al2O3. .................................................................. 41
4.9 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 3.1%ZrO2-Al2O3 where Al2O3 is pink (left) and ZrO2 isgreen (right). ........................................................................................................................... 42
NASA/TM—2012-216020 ix
4.10 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results overlaid on scanning electron microscopy image for3.1%ZrO2-Al2O3 where Al2O3 is pink (left) and ZrO2 is green (right). ............... 42
4.11 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)mapping results for 3.1%ZrO2-Al2O3 overlaid on scanning electronmicroscopy image where Al2O3 is pink (left) and ZrO2 is green (right)........... 43
4.12 Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)Quantitative Results for 3.1%ZrO2-Al2O3 ..................................................................... 44
4.13 Expected BET surface areas of (a) 15%Co/9.7%TiO2-Al2O3, (b)15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-Al2O3 ......................................... 46
4.14 BET surface area comparison of (a) 15%Co/9.7%TiO2-Al2O3 , (b)15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-Al2O3, (d) 15%Co/Al2O3HP14/150, (e) 15%Co/Al2O3 SBA150, and (f) 15%Co/Al2O3 SBA200. ............ 47
4.15 Barrett Joyner Halenda (BJH) average pore radius adsorption data for (a)15%Co/9.7%TiO2-Al2O3 , (b) 15%Co/3.0%La2O3-Al2O3, (c)15%Co/3.1%ZrO2-Al2O3, (d) 15%Co/Al2O3 HP14/150, (e) 15%Co/Al2O3SBA150, and (f) 15%Co/Al2O3 SBA200. ....................................................................... 48
4.16 Barrett Joyner Halenda (BJH) average pore volume desorption data for (a)15%Co/9.7%TiO2-Al2O3 , (b) 15%Co/3.0%La2O3-Al2O3, (c)15%Co/3.1%ZrO2-Al2O3, (d) 15%Co/Al2O3 HP14/150, (e) 15%Co/Al2O3SBA150, and (f) 15%Co/Al2O3 SBA200. ....................................................................... 48
4.22 H-TPR (solid) and (dashed) H-TPR-AR profiles of, moving upward, (a)15%Co/9.7%TiO2-Al2O3; (b) 15%Co/3.0%La2O3-Al2O3; and (c)15%Co/3.1%ZrO2-Al2O3. Reduction was carried out in hydrogen for 10hours at 350oC. ....................................................................................................................... 55
4.24 Temperature programmed desorption data for comparison of all catalystswith their corresponding reference catalysts. ........................................................... 59
4.25 Extent of reduction results from pulse reoxidation testing for each catalyst incomparison to its corresponding reference catalyst. .............................................. 60
4.26 Pulse reoxidation cluster size data, which is corrected for extent of reductionfor each catalyst in comparison to it’s reference catalysts. ................................... 61
4.27 Corrected dispersion data for each catalyst in comparison to its referencecatalysts ..................................................................................................................................... 62
4.28 X-ray diffraction profile of 15%Co/9.7%TiO2-Al2O3 (a) and 9.7%TiO2-Al2O3(b). ............................................................................................................................................... 65
4.29 X-ray diffraction profiles of 15%Co/3.0%La2O3-Al2O3 (a) and 3.0%La2O3-Al2O3(b). ............................................................................................................................................... 66
4.34 Adsorption pore size distribution of (left) 9.7%TiO2-Al2O3 and (right)15%Co/9.7%TiO2-Al2O3. .................................................................................................... 80
4.35 Desorption pore size distributions of 9.7%TiO2-Al2O3(left) and15%Co/9.7%TiO2-Al2O3(right) ........................................................................................ 80
NASA/TM—2012-216020 xi
4.36 Adsorption pore size distribution of 3.0%La2O3-Al2O3 (left) and15%Co/3.0%La2O3-Al2O3 (right) .................................................................................... 80
4.37 Desorption pore size distribution of 3.0%La2O3-Al2O3 (left) and15%Co/3.0%La2O3-Al2O3 (right) ..................................................................................... 81
4.38 Adsorption pore size distribution of 3.1%ZrO2-Al2O3 (left) and15%Co/3.1%ZrO2-Al2O3 (right) ...................................................................................... 81
4.40 Adsorption pore size distribution of Al2O3 HP14/150 (a)& (b) and15%Co/Al2O3 HP14/150 (c) & (d). ................................................................................ 82
4.41 Desorption pore size distribution of Al2O3 HP14/150 (a)& (b) and15%Co/Al2O3 HP14/150 (c) & (d). ................................................................................ 83
4.42 Adsorption pore size distributions of (a) & (b): SBA 150 Al2O3 and (c) & (d):15%Co/SBA 150 Al2O3. ....................................................................................................... 84
4.43 Desorption pore size distributions of (a) & (b): SBA 150 Al2O3 and (c) & (d):15%Co/SBA 150 Al2O3. ....................................................................................................... 85
4.44 Adsorption pore size distributions of (a) & (b): Al2O3.SBA 200 and (c) & (d):15%Co/Al2O3.SBA 200 ........................................................................................................ 86
4.45 Desorption pore size distributions of (a) & (b): Al2O3.SBA 200 and (c) & (d):15%Co/Al2O3.SBA 200 ........................................................................................................ 87
NASA/TM—2012-216020 xii
CHAPTER I
INTRODUCTIONCHAPTER 1
1.1 Fischer Tropsch Synthesis
The Fischer-Tropsch process has three distinct steps: gasification, synthesis,
and product upgrade. The gasification step produces syngas (hydrogen and carbon
monoxide) from many carbon resources. The synthesis involves the conversion of
syngas to syn-crude. Product upgrade processes the syn-crude and separates it into
useable liquid fuels. The synthesis step can be optimized to increase yields and
reduce energy inputs into the overall process.
1.2 Fischer-Tropsch Cobalt Catalysts
Cobalt-based Fischer-Tropsch catalysts are typically cobalt oxides on various
ceramic supports (e.g., alumina, silica, titanium oxide, etc). The support adds
mechanical stability, as well as an increased surface area for dispersion of the active
metal, cobalt. Support modification can change the interaction of the cobalt with the
support in order to increase activity and selectivity to the desired product. The
supported cobalt catalyst requires a reduction treatment to convert cobalt oxides to
metallic cobalt which catalyzes F-T synthesis reactions. This reduction step
NASA/TM—2012-216020 1
becomes a very important step in the process and the development of the cobalt
based catalysts and potentially their modified supports.
1.3 Objectives and Hypothesis
The objective of the presented work is to evaluate the effect of a few
structural promoters in the form of oxides on alumina supported cobalt catalysts.
Of these structural promoters, the modification of alumina with titanium,
lanthanum, and zirconium were the focus. The presence of these oxides on the
support was investigated using a wide range of characterization techniques such as
SEM, nitrogen adsorption, x-ray diffraction (XRD), temperature programmed
reduction (TPR), temperature programmed reduction after reduction (TPR-AR), and
hydrogen chemisorptions/pulse reoxidation. These characterization techniques
were used as a screening mechanism for the variety of cobalt/mixed oxide catalysts.
Since the physical properties of these mixed oxide supports were inherently
different, three different baseline alumina supported catalysts were used as
reference catalysts. The presence of these structural promoters in the form of
oxides could modify the alumina support properties and weaken the interaction of
cobalt with alumina. An optimized weakened interaction could lead to the most
advantageous cobalt dispersion, particle size, and reducibility. This optimization
could maintain the high number of active sites, while minimizing cobalt aluminate
formation. The hypothesis is that the presence of titanium oxide, lanthanum oxide,
NASA/TM—2012-216020 2
and zirconium oxide will reduce the interaction between cobalt and the alumina
support.
1.4 Outline
Chapter II will provide an extensive literature survey and provide
background information on the topic of interest. Previous research is presented in
the area of titanium, lanthanum, and zirconium as both a structural and reduction
promoter on alumina and similar ceramic supports. Chapter III provides the
experimental procedures used in making the catalysts and the characterization of
the catalysts in this research. Chapter IV provides all characterization results on the
three improved catalysts, as well as the three reference catalysts, as well as
discussion and conclusions.
NASA/TM—2012-216020 3
CHAPTER II
BACKGROUND OF THE STUDYCHAPTER 2
2.1 Principles of Cobalt Fischer-Tropsch Catalysis
Fischer-Tropsch (F-T) synthesis takes gaseous hydrogen and carbon
monoxide and converts it into various hydrocarbon chain length product
distributions. It is considered a network of parallel and consecutive reactions which
take place on the surface and in the pores of catalysts5. Cobalt based catalysts have
good activity and selectivity6 and are known to produce high molecular weight
paraffinic waxes that can be hydrocracked to produce lubricants and diesel fuels,
which make them of high interest in F-T synthesis. The use of supported cobalt
catalysts in F-T synthesis has led to complex development of catalyst design.
Changes in support, support modifications/promotion, cobalt loading and additional
promoter metals have been shown to change the performance of these catalysts
drastically.
NASA/TM—2012-216020 4
2.2 Fischer-Tropsch Cobalt Catalyst Oxide Support Interaction Effect
In order to increase the cobalt active sites, the cobalt metal is dispersed as
clusters on high surface area supports, typically oxides or mixed oxides. The oxides
are of particular interest because they are a highly porous structure, which is
theoretically inert in the F-T reaction. The physical properties of the supports help
increase surface area and distribute the cobalt metal clusters over the surface of the
support. The dispersion of active metal on the catalyst support is dependent on the
interaction of the support with the active metal, in this case cobalt. If the support
has a strong interaction with cobalt, it is likely that the cobalt will be highly
dispersed in small clusters on the surface of the oxide support. The support type
and physical properties determine the number of active sites after reduction, and
also influence the percentage of cobalt oxide species that can be reduced7.
Many different oxides can be used as the support for cobalt catalysts in
Fischer-Tropsch synthesis. A few of the most common supports are SiO2, TiO2, and
Al2O3; each of which has its own advantages and disadvantages. Al2O3 has a strong
interaction with cobalt, TiO2 has a moderate interaction with cobalt, and SiO2 has a
weak interaction with cobalt8. Khodakov et al. show that the second reduction step
(CoO to Coo) is strongly influenced by the cobalt particle size, such that smaller and
more interacting particles (6nm) are more difficult to reduce than larger cluster
sizes (20-70nm) in studying SiO2 as the support9. These interactions play an
important role in the activity of the catalysts because of the tendency to form cobalt-
oxide complexes such as cobalt aluminate, cobalt silicate, or cobalt titanate. For
example, TiO2 has a strong metal-support interaction with cobalt, which makes TiO2
NASA/TM—2012-216020 5
catalysts difficult to reduce3. This has been attributed to the strong cobalt-oxide
interaction with the support and shifts the reduction temperature to much higher
than preferred3. The properties of the support have been shown to play a role in the
F-T kinetics8 and have been linked to the catalytic performance. It is in the best
interest of researchers to determine a means to reduce the use of promoters and
expensive metals, while at the same time increasing the performance of these cobalt
catalysts3.
Alumina tends to be favorable due to the mechanical properties10,
particularly in applications such as continuously stirred tank reactors11 where
agitation is present. Previous research has shown that alumina supported cobalt
does not completely reduce from cobalt oxide to cobalt metal because of the high
cobalt metal-support interactions7. This high interaction of alumina with cobalt11
results in the tendency to form cobalt aluminate species, which is likely inactive
cobalt. With cobalt aluminate formation, usually cobalt loadings need to be higher
than 20% in order to achieve desired activity.
2.3 Cobalt Metal Loading and Size
Increasing the cobalt loading on alumina has been shown to decrease the
reduction temperature in both unpromoted and promoted catalysts. The cobalt
cluster size has been linked to the interaction of cobalt with the support8, where
having a very strong interaction of cobalt with the support results in small cluster
sizes and a very dispersed active metal on the surface of the catalyst.
NASA/TM—2012-216020 6
Wang and Chen 12 showed that when low loadings of cobalt are used, cobalt
aluminate is favored. Cobalt aluminate can only be reduced at high reduction
temperatures, which aren’t feasible because of cobalt agglomeration. Their research
also indicated that higher loadings of cobalt resulted with Co3O4 crystallites that
were easier to reduce, showing a single broadened TPR peak12. It has also been
shown that the cluster size and the support effect play a role in the F-T kinetics8 and
that the kinetic reaction orders vary based on the size of the metal crystallites13.
2.4 Promoter Metals
One key concern with handling cobalt based catalysts is that the active form
of catalyst is in a reduced state, metallic cobalt, which oxidizes readily in air.
Therefore, cobalt catalysts require a reduction step before F-T synthesis can occur.
Since some cobalt-support interactions are high, it becomes difficult to fully reduce
the available cobalt metal on the surface of the support. Noble metals, such as Ru,
Pt, and Re are commonly added as reduction promoters in order to produce more
cobalt metal surface sites by facilitating reduction of cobalt species that interact
with the support14. In most cases, the application of a promoter is used in order to
enhance the reduction of cobalt oxide (CoO) to cobalt metal (Coo). The addition of
the platinum to the surface of a Co/Al2O3 catalyst has been proven to decrease the
reduction temperature of the catalysts, most likely due to hydrogen spillover from
the metallic promoter7. It has been found that the addition of Pt promoter to the
NASA/TM—2012-216020 7
cobalt catalyst not only enhances the reducibility of the clusters, it also causes a
small decrease in cobalt cluster size7.
2.5 Preparation Method
In cobalt catalyst synthesis, there are many different ways to load the cobalt
onto the ceramic support. Some of the common methods are: impregnation, co-
precipitation, sol-gel, chemical vapor deposition, and plasma15. Of these ways,
impregnation is one of the most popular, having two common impregnation
methods: incipient wetness impregnation (IWI) and slurry phase impregnation.
Previous research has shown that incipient wetness impregnation procedures have
been found to produce a wider range of cluster sizes than the slurry phase
impregnation procedure7. It is widely accepted that TiO2 as a support leads to a
stronger cobalt- support interaction than SiO216, but it was found that the addition of
TiO2 to SiO2 only improved catalyst performance when applying 3 out of 4
preparation techniques16. Improvements were found in precipitation,
impregnation, and hydrolysis-reflux, although not in sol-gel methods16. It has been
shown in other instances that preparation procedure can change the outcome of the
results, which should be considered in the evaluation of the structural promoter
discussion in this work.
NASA/TM—2012-216020 8
2.6 Pore Size
The pore size of the ceramic support is important because it allows for the
diffusion of reagents and products inside the catalyst channels and pores. This
means that the pore size could limit the reaction, or possibly favor one reaction over
another in F-T synthesis. Khodakov et. al 15 express that support pore size could
affect the diffusion and capillary condensation of products in these pores and that
narrow pores are more likely to be filled with liquid products than the wider pores
in the catalyst. Liu et all found that the pore size of SiO2 supports greatly influence
the activity of the catalyst in a 0.5L CSTR17. Their research showed an optimum
average pore size of 10nm in order to increase the activity17.
2.7 Mixed Oxide Supports
Much research has been done in the area of mixed oxide supports. The focus
has been on SiO2 and TiO2 supports, however some research has been done on
Al2O3.
2.7.1 Titanium oxide (TiO2)
Titanium oxide has been used as a ceramic support alone in F-T catalysis. It
has a high interaction with cobalt, although not as high as alumina. In one study, the
addition of TiO2 to SiO2 catalyst supports had drastically affected the structure and
catalytic performance of the cobalt F-T catalysts16, likely by modification of the
interaction between cobalt and silica. Preliminary results indicate that
NASA/TM—2012-216020 9
impregnation method resulted in the highest activity and selectivity for C5+16. The
preparation procedures had drastic affected the reduction profiles of these catalyst,
through addition of a third (small) peak, which the authors attributed to an
interaction compound16. This indicates that TiO2 promoted catalysts were more
difficult to reduce16, which is expected since TiO2 has a stronger interaction with
cobalt. The cobalt particle size decreased upon addition of TiO216, which was
expected since smaller particles are also considered to interact more strongly with
the support and be more difficult to reduce. The XRD profile showed slightly
broader peaks for Co3O4 crystallites on the TiO2 promoted catalysts16 and using the
Scherrer equation, the silica only supported catalyst was found to have the largest
Co3O4 crystallite sizes16. These results confirm that the stronger interaction
between titania and cobalt produce higher dispersed catalysts with smaller cobalt
particles and that the addition of TiO2 to silica support can provide an added benefit
to the catalyst surface composition.
Wan et al. found a clear change in morphology from pure Al2O3 by the
addition of TiO2 to the surface by the sol-gel method, where with increasing TiO2
weight percent, XRD profiles indicate an increase in peak intensity in anatase18,
which signifies an increase in TiO2 cluster size. It is important that the oxide dopant
is well dispersed for a uniform structure and more predictable catalytic activity.
NASA/TM—2012-216020 10
2.7.2 Lanthanum Oxide (La2O3)
Lanthanum has also been known to increase activity in Fischer-Tropsch
synthesis. A study by Vada et al. concludes that La3+ increased the overall activity
and chain-growth probability when a low loading of La3+ (La/Co=0.05) was present
on a Co/Al2O3 catalyst, however methane production increased as well19. The
authors of this study also found that the catalyst activity decreased with higher
loadings of La (La/Co )19. Since methane is not a desirable product in F-T
synthesis, these results indicate that there is likely a positive and negative effect of
lanthanum as a structural promoter on F-T cobalt alumina supports, which may be
dependent on the amount of cobalt loaded on the support. There is significant need
to determine the optimal loading of lanthanum on alumina, as well as the
preparation procedure. Further investigation is necessary in order to determine the
physical and kinetic influence of lanthanum as an oxide support modification.
The preparation procedure for loading lanthanum on alumina was studied by
Ledford et al. and found that there was only a significant change in reducibility and
cobalt metal dispersion when La3+ was impregnated before cobalt loading20. The
authors also determined that higher La3+ loadings (La/Al>0.026) resulted in
formation of a La-Co mixed oxide and enhanced the dispersion of cobalt on the La-
Al2O320. The increase in dispersion showed little correlation to the reducibility in
this study and the effect of increased dispersion on reducibility should be further
explored.
NASA/TM—2012-216020 11
Cai et al. also compared preparation procedures by preparing La2O3-doped
alumina catalysts through impregnation and co-precipitation and found that the co-
precipitation method reduced more readily and resulted in higher F-T synthesis
activity and lower methane selectivity21. DRIFTS studies indicated that CO adsorbed
most easily on the co-precipitated support catalyst than on the SASOL commercial
support21.
Zhang et al. promoted titania with lanthanum nitrate at varying atomic ratios
of La/Ti, calcined the support to drive off the nitrate and proceeded to load 12%
cobalt to the La promoted Ti22. They found that as the La content increased, the
cluster size of Co3O4 decreased, while the percent reduction increased22. This was
not to be expected because typically, larger clusters reduce easier. It appears that as
the La loading increases, the reduction peak narrows and shifts to a lower
temperature. It is also interesting to note that the nitrate peak shifted to a higher
temperature and the area increased as the La loading increased, which was
explained as a possible stabilization of nitrate during the calcination process22. The
authors found that La inhibits nitrate degradation during calcination and may
require higher temperature calcination if used as a structural promoter. The study
completed x-ray diffraction spectroscopy and found little indication of change in
crystal size in varying La promotion levels22. In comparison of non-promoted
catalysts to La promoted catalysts, there is no indication of lanthanum crystallites in
the XRD profile22. This suggests that the lanthanum is in highly dispersed form22
and that it had little effect on the cobalt crystallite size. Using the Sherrer’s
equation, the crystalline size of Co3O4 was calculated and it indicated that the
NASA/TM—2012-216020 12
average particle size decreased with increasing lanthanum content, which results in
a much higher dispersion22. The introduction of lanthanum on a titania supported
cobalt catalyst could have a positive effect on the dispersion, which has been linked
to the activity of the catalyst.
2.7.3 Zirconium Oxide (ZrO2)
A significant amount of research has been done in the addition of ZrO2 to SiO2
and TiO2, with little on the modification of Al2O3. The effect of zirconium
modification to all supports seems to be controversial; however there is also
differences in the preparation of these catalysts among authors. As previously
discussed, it is difficult to separate the effect of preparation procedures, cobalt metal
loading, and calcination temperatures to the zirconium loading effect.
Some research has shown higher activity and C5+ selectivity23 in Fischer-
Tropsch synthesis. Ali et al. 24suggested that the promotion of zirconium on silica
created an active interface with Co, which facilitates CO dissociation and thus
increasing the activity. Rohr et al. 25 concluded that the modification of ZrO2 on
Co/Al2O3 increased the activity and selectivity to heavier hydrocarbons, which was
attributed to changes in surface coverage of reactive intermediates.
Some research has shown a weaker cobalt-zirconium interaction on Co/SiO2
catalysts, which led to an increase in reducibility26. Other research indicates that
there is no decrease in reduction for the addition of zirconia to Co/SiO2 catalysts27.
Moradi et al. 23 also found that the addition of zirconia to silica favors reduction at a
NASA/TM—2012-216020 13
lower temperature concluding that the cobalt-silica interaction is replaced by the
cobalt-zirconia interaction. The deposition of zirconia on silica support was proven
to prevent the formation of cobalt silicate,27 which also may mean that the cobalt-
silica interaction is affected by the zirconia being present. Michalak et al, found that
the addition of ZrO2 had no impact on the surface area of the catalyst, but it
decreased the extent of reduction for Co/Al2O3 catalysts27. This may suggest that
ZrO2 inhibits the reduction of cobalt catalysts, supported on alumina.
XRD signals of both amorphous alumina and the addition of ZrO2 to the
alumina did not indicate any changes in signal, which leads us to believe that there
was a strong interaction between ZrO2 and the oxide composite.28 This strong
interaction likely resulted in highly dispersed ZiO2 on alumina.
Xiong et al. synthesized ZrO2-Al2O3 through impregnation of zirconium to a
Al2O3 and found that the cobalt oxide crystallite size increased with
increasing zirconium, while the zirconia inhibited formation of cobalt aluminate in
those catalysts29. Xiong also found that an increase in zirconia decreased the
methane selectivity, increased CO hydrogen activity, and C5+ selectivity in Fischer-
Tropsch synthesis.
2.8 Brunauer Emmet Teller (BET) Surface Area Measurements
Brunauer, Emmet, Teller (BET)4 surface area measurements are important
for Fischer-Tropsch catalysts because the results provide surface area data, which
are necessary to determine the available surface area for the reaction to take place.
NASA/TM—2012-216020 14
The surface area of granulated powders is measured by determining the quantity of
gas that adsorbs as a single layer of molecules, which is completed near the boiling
point of the adsorbate gas. At the boiling point, the area covered by each gas
molecule is known within minimal error and the sample surface area is calculated
directly from the number of absorbed molecules, the conditions, and the area
occupied by each molecule. In most instances, 30% nitrogen in helium mixture is
used at atmospheric pressure and at liquid nitrogen temperature. The adsorption of
gas on a solid surface is described by the following equation4 :
1
1 =
1
+
1
=
=
= ( )
=
The surface area (S) of the sample is determined by the monolayer of
absorbed gas volume (Vm) at standard temperature and pressure, given in the
following equation:
=
=
==
Applicable BET surface area data and BJH adsorption data of Fischer-Tropsch
cobalt catalysts available in literature is presented in Table 2.1. This data provides a
NASA/TM—2012-216020 15
means for comparison of data collected in the work presented in this thesis to other
literature available.
Table 2.1: BET Data from Literature
BETSurface
Area(m2/g)
Single Pointadsorption
average porevolume (cm3/g)
Single pointadsorption average
pore radius (nm)
Ref.
15%Co/Al2O3 144 0.38 5.25 29
15%Co/1%Zr/Al2O3 159 0.4 5.0 29
15%Co/5%Zr/Al2O3 133 0.35 5.2 29
15%Co/9%Zr/Al2O3 123 0.31 5.1 29
Al2O3 136 27
8.5%ZrO2/Al2O3 120 27
10%Co/Al2O3 103 27
10%Co/ZrO2/Al2O3 75 27
TiO2 49.5 27
8.5%ZrO2/TiO2 47.6 27
10%Co/TiO2 38.3 27
10%Co/ZrO2/TiO2 38.0 27
**Blank spaces were left when data was unavailable
NASA/TM—2012-216020 16
2.9 Temperature Programmed Reduction and Temperature ProgrammedReduction After Reduction
Since Fischer-Tropsch cobalt catalysts oxidize readily in air, it is critical that
the catalysts are reduced before introducing them to the reaction. If reduction is not
completed at the optimum conditions, the catalyst may sinter and/or agglomerate
during the process. Thermodynamics determine the best conditions at which a
catalyst can be reduced, but are only useful if the catalyst particles are equivalent30 ,
meaning that the cobalt particle sizes are uniform and is typically not the case with
Fischer-Tropsch catalysts.
TPR provides useful information on the temperature that is needed for the
complete reduction of the catalyst 30. However, this temperature is often not the
optimum condition for reduction, many other factors are considered for this
determination. The area under the TPR curve represents the total hydrogen
consumption and is commonly expressed in moles of H2 consumer per mole of metal
atoms (H2/M)30. Most frequently, TPR profiles are interpreted on a qualitative basis
and not a quantitative.
During TPR, the metal oxide (cobalt oxide) reacts with hydrogen to form
pure metal (cobalt metal). This reaction is also known as reduction. During the
TPR, argon is used as the carrier gas because it has a very low relative thermal
conductivity. The argon is blended in a fixed proportion with hydrogen, the
reduction gas, which has a much higher thermal conductivity. The gas mixture flows
through the analyzer, the sample, and then the detector. A baseline reading is
established by the detector when the initial H2/Ar gas flows over the sample. This
NASA/TM—2012-216020 17
occurs at a low enough temperature that no reduction has begun. As the
temperature is increased at a fixed ramp rate, the hydrogen atoms begin to react
with the sample. This reaction produces H2O molecules, which are removed from
the gas stream using a cold trap. The production of H2O results in a decrease in the
amount of hydrogen, thus shifting the total gas thermal conductivity towards the
argon’s thermal conductivity. As previously mentioned, argon has a lower thermal
conductivity than hydrogen, so the total gas thermal conductivity decreases. The
signal the instrument records is the electrical demand, also known as the detector
signal. This demand is described as the amount of electricity it takes to keep the
TCD at a constant filament temperature. So, as the total gas thermal conductivity
decreases, the flowing gas removes heat from the filament more slowly, therefore it
requires less electricity to maintain the filament temperature.
2.10 Hydrogen Chemisorption and Pulse Reoxidation
Hydrogen chemisorption and pulse reoxidation provide useful information
about the catalyst’s active site density, dispersion, cluster size, and reducibility. The
hydrogen temperature programmed desorption (TPD) provides data on the active
site density and is also used to calculated uncorrected dispersion. These
calculations are done under the assumption that all of the cobalt was reduced and
that one hydrogen atom attaches to one surface cobalt metal atom. Since it was not
completely reduced, the use of pulse reoxidation becomes important. The pulse
reoxidation date is used to calculate the extent of reduction, which is then used to
determine the corrected dispersion and cluster size. The assumption in this
NASA/TM—2012-216020 18
calculation is that for every two oxygen molecules consumed, there are three moles
of bulk cobalt metal atoms previously reduced.
NASA/TM—2012-216020 19
Tabl
e 2.
2: H
ydro
gen
Chem
isor
ptio
n Da
ta fr
om L
itera
ture
Cata
lyst
Des
crip
tion
H2
deso
rbed
per g
cat
Unco
rrec
ted
Dis
pers
ion
(%)
Unco
rrec
ted
Diam
eter
(nm
)m
oles
O2
cons
umed
per g
cat
%Re
duct
ion
Corr
ecte
dD
ispe
rsio
n(%
)
Corr
ecte
dDi
amet
er(n
m)
Ref .
10%
Co-0
%M
n-Ti
O27.
1611
.831
10%
Co-0
.03%
Mn-
TiO2
6.03
9.7
31
10%
Co-0
.32%
Mn-
TiO2
7.16
10.4
31
10%
Co-0
.96%
Mn-
TiO2
2.50
9.2
31
10%
Co-3
.43%
Mn-
TiO2
1.77
9.6
31
15%
Co-S
iO2
CS
7.4
91.1
17.2
16
15%
Co-T
iO2/
SiO2
CTSP
8.8
64.6
14.6
16
NASA/TM—2012-216020 20
Cata
lyst
Des
crip
tion
H2
deso
rbed
per g
cat
Unco
rrec
ted
Dis
pers
ion
(%)
Unco
rrec
ted
Diam
eter
(nm
)m
oles
O2
cons
umed
per g
cat
%Re
duct
ion
Corr
ecte
dD
ispe
rsio
n(%
)
Corr
ecte
dDi
amet
er(n
m)
Ref .
15%
Co-T
iO2/
SiO2
CTSI
8.7
69.9
14.7
16
15%
Co-T
iO2/
SiO2
CTSH
R
8.6
61.2
14.8
16
15%
Co-T
iO2/
SiO2
CTSS
G
10.2
35.4
12.6
16
12%
Co-T
i-0La
81.6
615
.722
12%
Co-T
i-1/2
000L
a82
.90
15.8
22
12%
Co-T
i-1/2
00La
84.1
213
.922
12%
Co-T
i-1/1
20La
87.2
512
.122
12%
Co-T
i-1/2
0La
88.7
612
.822
15%
Co/A
l 2O3
88.4
6.94
14.9
898.
452
.68
13.2
7.9
29
15%
Co/0
.5%
Zr/A
l 2O3
74.0
5.82
17.7
959.
153
.24
10.4
10.0
29
15%
Co/1
%Zr
/Al 2O
374
.25.
8317
.591
0.2
53.3
710
.99.
329
NASA/TM—2012-216020 21
Cata
lyst
Des
crip
tion
H2
deso
rbed
per g
cat
Unco
rrec
ted
Dis
pers
ion
(%)
Unco
rrec
ted
Diam
eter
(nm
)m
oles
O2
cons
umed
per g
cat
%Re
duct
ion
Corr
ecte
dD
ispe
rsio
n(%
)
Corr
ecte
dDi
amet
er(n
m)
Ref .
15%
Co/5
%Zr
/Al 2O
310
2.5
8.05
12.8
855.
850
.18
16.0
6.4
29
15%
Co/9
%Zr
/Al 2O
399
.27.
0114
.710
05.8
58.9
511
.98.
729
15%
Co/1
5%Zr
/Al 2O
3
100.
07.
7313
.410
20.4
59.9
115
.18.
029
10%
Co/A
l 2O3
0.4
6727
027
10%
Co/8
.5%
ZrO 2
/Al
2O3
1.1
6110
127
**Bl
ank
spac
es w
ere
left
whe
n da
ta w
as u
nava
ilabl
e
NASA/TM—2012-216020 22
2.11 X-ray Diffraction
X-ray diffraction provides a way to estimate the cobalt oxide crystallite size.
29. and spinel cobalt oxide is found at 31.3,
36.9, 45.1, 59.4, and 65.432. Table 2.3 provides XRD data from literature for a
baseline comparison to the data presented in this study.
Table 2.3: XRD Data from Literature
Catalyst Average diameter ofCo3O4 domains (nm)
Ref.
15%Co/Al2O3 18.3 29
15%Co/0.5%Zr/Al2O3 18.2 29
15%Co/1%Zr/Al2O3 19.1 29
15%Co/5%Zr/Al2O3 19.3 29
15%Co/9%Zr/Al2O3 18.4 29
15%Co/15%Zr/Al2O3 20.8 29
NASA/TM—2012-216020 23
CHAPTER III
EXPERIMENTAL PROCEDURESCHAPTER 3
3.1 Catalyst Preparation
In this study, Puralox SCFa-140/L3 (Sasol), Puralox SCFa-200Zr3 (Sasol), and
Puralox TH 100/150 Ti10 (Sasol) were used as the explored catalyst supports.
Puralox HP14/150 Al, Catalox Al2O3 SBA200, and Catalox Al2O3 SBA150 were used
as the reference catalyst supports. These reference pairs were chosen because of
their surface area and pore size data collected: 9.7%TiO2-Al2O3 compared to both
Al2O3 SBA 150 and Al2O3 HF14/150, 3.1%ZrO2-Al2O3 compared to both Al2O3 SBA
150 and Al2O3 SBA 200, and 3.0%La2O3-Al2O3 compared to both Al2O3 SBA 150 and
Al2O3 SBA 200. All six supports are readily available for purchase from Sasol North
America. Table 3.1 shows the composition, surface area, loose bulk density, and
particle size distribution data provided by Sasol North America upon delivery of
these supports.
NASA/TM—2012-216020 24
Table 3.1: Catalyst support composition
Support Name Composition SurfaceArea
(m2/g)
Loose BulkDensity
(g/l)
Particle SizeDistribution
%
Puralox TH 100/150 Ti10 90.3%Al2O39.7%TiO2
135 0.33 <25 µm 50.6
<45 µm 82.5
<90 µm 100
Puralox SCFa-140/L3
Lot No: BD2186
97% Al2O3
3%La2O3
143 0.61 <25 µm 33
<45 µm 60
<90 µm 94.7
>150µm
0.1
Puralox SCFa-200 Zr3
Lot No: BD2801
96.9%Al2O3
3.1%ZrO2
196 0.67 <25 µm 49.2
<45 µm 82.1
<90 µm 100
>150µm
0
Puralox HP14/150 Al 100%Al2O3 150
Catalox Al2O3 SBA200 100%Al2O3 200
Catalox Al2O3 SBA150 100%Al2O3 150
Support calcination is necessary to drive off any water that may be absorbed
on the support from the atmosphere. The support calicnations were carried out
using a tubular reactor and a clamshell furnace. A Lindberg/Blue M control console
was used to set the parameters of the experiment and control the internal
temperature of the support throughout the calcination procedure.
NASA/TM—2012-216020 25
For all three catalyst supports, approximately 30 grams were loaded into the
reactor. The reactor was then loaded into the furnace, and the air supply turned on
at a flow rate of approximately 2.0 L/min set on a rotameter. The source of the air
used in the experiment was a gas cylinder of zero air. The controller was set to
ramp at 2°C/min from room temperature to 400°C and then it was held at 400oC for
four hours. The sample was then kept under air flow until reaching room
temperature and then removed from the tubular reactor and stored in an oven at
100oC until the first cobalt loading began. Figure 3.1 shows the diagram of the
calcination reactor, a plug flow reactor (PFR). The air flows from the top of the
reactor over the catalyst bed and out the bottom of the reactor, with a thermocouple
located in the middle of the catalyst bed.
NASA/TM—2012-216020 26
Figure 3.1: Calcination reactor drawing
Impregnation is one of the many different methods for synthesizing Fischer-
Tropsch (F-T) catalysts, which can be done through incipient wetness impregnation
(IWI) or slurry impregnation. Incipient wetness impregnation uses a loading
solution that is equal to the exact volume of pores in the support, while slurry
impregnation requires the final loading solution to be equal to 2.5 times the total
pore volume of the support. Previous research has shown that incipient wetness
impregnation procedures have been found to produce a wider range of cluster sizes
than the slurry phase impregnation procedure7. Because of the wide range of
NASA/TM—2012-216020 27
cluster sizes in IWI procedures, slurry phase impregnation was chosen for this
study.
The catalyst was prepared using slurry impregnation and two separate
loadings of cobalt were made using cobalt nitrate and water. The first loading
required half of the cobalt nitrate needed for 15% by weight loading. This cobalt
nitrate was dissolved in de-ionized water so that the total volume was equivalent to
2.5 times the total pore volume for the specified support. The solution was then
added (drop wise via a burette and a rotating round bottom flask) to the catalyst
support until approximately ¼ of the solution was dispensed. At this time, the
round bottom flask was removed from the rotating mechanism and thoroughly
mixed by shaking and scraping the walls of the round bottom flask. This process
was continued until the entire solution was added to the support. The round
bottom flask was then transferred to the rotary evaporator where the vacuum was
controlled to ensure very slow drying of the catalyst. The second loading of cobalt
was completed using the same methods.
After completion of both cobalt nitrate loadings, the nitrate must be driven
off through catalyst calcination. This calcination was completed using the same
furnace and control system previously mentioned in the support calcination
description; however a separate tubular reactor was used to avoid contamination.
Zero air flow was set to approximately 2L/min and the temperature controller was
set to ramp up at a rate of 2oC/min from room temperature to 350oC and hold at
350oC for 4 hours. After completion of calcination, the air flow was maintained until
proper cool down of the catalyst.
NASA/TM—2012-216020 28
3.2 Catalyst Characterization
A number of catalyst characterization techniques were used including:
Scanning electron microscopy (SEM) and electron dispersed spectroscopy(EDS),
Brunauer Emmet Teller (BET), Barrett Joyner Halenda (BJH), temperature
programmed reduction (TPR), temperature programmed reduction after reduction
(TPR-AR), hydrogen chemisorption with pulse reoxidation, and x-ray diffraction
(XRD).
3.2.1 Scanning Electron Microscopy and Electron Dispersed Spectroscopy
Scanning electron microscopy (SEM) images were gathered on only doped
alumina supports using a Hitachi S-3000N equipped with an EDAX detector for
electron dispersed spectroscopy (EDS) measurements. The SEM was set to 25kV for
imaging and the EDAX working distance was set to 15mm before measurements
were taken. SEM-EDS measurements were gathered on each of the doped alumina
supports in order to better understand the surface morphology of the dopants. The
samples were prepared and mounted on copper tape before measurements were
taken.
NASA/TM—2012-216020 29
3.2.2 Brunauer Emmet Teller (BET) Surface Area Measurements and BarrettJoyner Halenda (BJH) Pore Size Distributions
BET (Brunauer, Emmet, and Teller4) and BJH (Barrett Joyner Halenda 33)
measurements were conducted on all of the supports, as well as the catalysts to
determine the loss of surface area after loading cobalt metal. These measurements
were conducted using a Micromeritics Tri-Star system. Approximately 0.5 grams of
sample was prepared by slowly ramping to 160oC and evacuating to 50mTorr. This
preparation step was completed in order to remove any water or other
contaminants on the surface of the catalyst or support. The BET surface area
measurements were completed with nitrogen and argon as the adsorption gases.
3.2.3 Temperature Programmed Reduction (TPR)
Temperature programmed reduction (TPR) profiles were obtained for each
of the calcined catalysts using a Zeton Altamira AMI-200 unit. Each sample was
loaded into a Zeton Altamira sample tube with a target mass of 0.1 grams. The
sample tubes were installed on the instrument and then set to undergo argon
pretreatment. The samples were heated to 350oC at a rate of 10oC/min under argon
flow of 30cm3/min in order to remove any residual water or nitrate. After argon
pretreatment, the sample was cooled to 50oC and held under continuous flow of
argon for 15 minutes. The flow was then switched from pure argon to 10%
hydrogen in argon (remaining at a constant 30 cm3/min). At this point the TCD
NASA/TM—2012-216020 30
signal started recording and the ramp rate was set to 10oC/min from 50oC to 1100oC
and the sample was held at 1100oC for a minimum of 30 minutes.
3.2.4 Temperature Programmed Reduction After Reduction
Temperature programmed reduction after reduction (TPR-AR) profiles were
obtained for all of the calcined catalysts using a Zeton-Altamira AMI-200 unit. The
first step was argon pretreatment step, which was mentioned previously in the TPR
section. After argon pretreatment, the sample was reduced under 30cm3/min flow
of 33% hydrogen in argon at a ramp rate of 1oC/min from room temperature up to
350oC and held for 10 hours. The flow was then switched to 10% hydrogen in argon
and the TCD began recording. The sample was then heated to 1100oC at a ramp rate
of 10oC/min.
3.2.5 Hydrogen Chemisorption with Pulse Reoxidation
Hydrogen chemisorption with pulse reoxidation measurements were
performed on the calcined catalysts using a Zeton Altamira AMI-200 unit, which
incorporates a thermal conductivity detector. Each sample weight was
approximately 0.22 grams. The catalyst was loaded via a sample tube and activated
at 350oC for 10 hours using a flow of 33% hydrogen in argon and then cooled under
hydrogen flow to 100oC. The sample was held at 100oC, while switching the flow to
pure argon in order to prevent physisorption of weakly bound species. The sample
was then slowly increased to activation temperature and held under flowing argon
NASA/TM—2012-216020 31
to desorb the remaining chemisorbed hydrogen and the TCD signal returned to
baseline. This TPD spectrum was integrated in order to find the number of
hydrogen moles desorbed in comparison to the area of the calibrated hydrogen
pulse’s peaks. The hydrogen TPD results are then used for calculating uncorrected
dispersion.
The same sample was reoxidized by injecting pulses of pure oxygen in helium
in reference to pure helium gas at the activation temperature. The number of moles
of oxygen consumed by the sample was determined by integration of the peaks and
using the same calibration method for hydrogen chemisorption. Assuming that all
of the Coo reoxidized to Co3O4, the percentage reduction was calculated. The
uncorrected dispersions are based on the assumption of complete reduction, where
the corrected dispersions include the percentage of reduced cobalt. The number of
Coo moles on the surface is determined by the number of hydrogen desorbed during
TPD and the total number of moles of Coo in the sample is the preparation target
weight percent of cobalt.
3.2.6 X-ray Diffraction
X-ray diffraction profiles were obtained for each of the calcined catalyst using
a Philips X’Pert unit. A long range scan was ran from 15 to 80o with 0.02o steps at 5
seconds/step. In order to quantify the average Co3O4 cluster sizes using the
o, which represents (3 1 1), a shorter range scan was
NASA/TM—2012-216020 32
also made. The shorter scan range was from 30 to 45o with 0.01o steps at 15
seconds/step.
NASA/TM—2012-216020 33
CHAPTER IV
CATALYST CHARACTERIZATION RESULTSCHAPTER 4
4.1 Scanning Electron Microscopy and Electron Dispersed Spectroscopy
Figure 4.1 shows the scanning electron microscopy-energy dispersive
spectroscopy (SEM-EDS) mapping results for both alumina and titanium for the
9.7%TiO2-Al2O3 support. The left image highlights the alumina in pink and the right
image highlights the titanium in green. The alumina appears to be the stronger
presence and highlights the particles shapes because of the concentration of
alumina being detected by SEM-EDS. The titanium appears to be highly dispersed
and have less concentration on the sample. This relationship was expected as the
9.7%TiO2-Al2O3 support contains less than 10% titanium. Since SEM-EDS has a
penetration of approximately 2µm, it is likely that most of the titanium is on or near
the surface of the support, which make it available to contribute to the F-T reaction.
NASA/TM—2012-216020 34
Figure 4.1: Scanning Electron Microscopy-Energy Dispersive Spectroscopy(SEM-EDS) mapping results for 9.7%TiO2-Al2O3 where Al2O3 is pink (left) and TiO2 isgreen (right).
Figure 4.2 shows the overlay of the SEM-EDS mapping results for both
alumina and titanium separately on the actual SEM image of the 9.7%TiO2-Al2O3
support. The left image highlights the alumina only (in pink) and the right image
highlights the titanium only (in green). As previously discussed, the alumina shows
a higher concentration and highlights the support particles structure because of this
higher concentration. The titanium appears to be uniformly dispersed over all
support particles in this image.
NASA/TM—2012-216020 35
Figure 4.2: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results overlayed on scanning electron microscopy image for
9.7%TiO2-Al2O3 where Al2O3 is pink (left) and TiO2 is green (right).
Figure 4.3 shows the overlay of both alumina and titanium SEM-EDS
mapping results on top of the actual SEM image of the 9.7%TiO2-Al2O3 support.
shows the SEM-EDS mapping overlay. This image puts previous conclusions into
perspective. It difficult it is to see the titanium among the alumina mapping results,
likely because of the highly dispersed titanium on the surface.
NASA/TM—2012-216020 36
Figure 4.3: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results for 9.7%TiO2-Al2O3 overlaid on scanning electron microscopy
image. Alumina is highlighted in pink and titania is highlighted in green.
Figure 4.4 and Table 4.1 show the SEM-EDS quantitative results for
9.7%TiO2-Al2O3. Figure 4.4 shows three distinct peaks: alumina, titanium, and the
copper tape used to mount the sample. Results indicate an atomic % of titanium as
8.17%, which is in order with the expected 9.7% of titanium.
NASA/TM—2012-216020 37
Figure 4.4: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) quantitative results for 9.7%TiO2-Al2O3 support.
Table 4.1: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) quantitative results for 9.7%TiO2-Al2O3
Element Wt% At% K-Ratio Z A F
AlK 86.35 91.83 0.7122 1.01 0.8162 1.0006
TiK 13.65 8.17 0.1128 0.9282 0.8904 1.0000
Figure 4.5 shows the scanning electron microscopy-energy dispersive
spectroscopy (SEM-EDS) mapping results for both alumina and lanthanum for the
3.0%La2O3-Al2O3 support. The left image highlights the alumina in pink and the
right image highlights the lanthanum in green. Again, alumina appears to be the
stronger presence and highlights the particles shapes because of the concentration
of alumina being detected by SEM-EDS. The lanthanum appears to be highly
dispersed and have less concentration on the sample, even less than the previously
discussed titanium. This relationship was expected as the 3.0%La2O3-Al2O3 support
NASA/TM—2012-216020 38
contains only 3% lanthanum. Since SEM-EDS has a penetration of approximately
2µm, it is likely that most of the lanthanum is on or near the surface of the support.
Figure 4.5: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results for 3.0%La2O3-Al2O3 where Al2O3 is pink (left) and La2O3 is
green (right)..
Figure 4.6 shows the SEM-EDS mapping results overlaid on the SEM image
for the 3.0%La2O3-Al2O3 support. The alumina is highlighted in pink on the left
image and shows the formation of the support particles. The lanthanum is shown
highlighted in green on the right and appears to have a high dispersion.
NASA/TM—2012-216020 39
Figure 4.6: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results overlaid on scanning electron microscopy image for
3.0%La2O3-Al2O3 where Al2O3 is pink (left) and La2O3 is green (right).
Figure 4.8 shows the SEM-EDS mapping of both alumina and lanthanum
overlaid on the SEM image. Similar to the titanium doped catalyst, the lanthanum is
difficult to see and is likely well dispersed throughout the alumina. It is highly likely
that the lanthanum is on or near the surface of the support and will play a role in the
catalytic reaction.
Figure 4.7: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results for 3.0%La2O3-Al2O3 overlaid on scanning electron
microscopy image (Al203 in pink and La2O3 in green).
NASA/TM—2012-216020 40
Figure 4.8: and Table 4.2 show the SEM-EDS quantitative results for the
3.0%La2O3-Al2O3 support. The figure displays peaks relating to alumina, lanthanum
and the copper tape used in mounting the sample. The quantitative results indicate
1.59% lanthanum, which is well within experimental error for the expected value of
3% lanthanum.
Figure 4.8: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) quantitative results for 3.0%La2O3-Al2O3.
Table 4.2Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS)quantitative results for 3.0%La2O3-Al2O3.
Element Wt% At% K-Ratio Z A F
AlK 92.33 98.41 0.7041 1.0116 0.7537 1.0002
LaL 7.67 1.59 0.0636 0.8007 1.0346 1.0000
NASA/TM—2012-216020 41
Figure 4.9: Scanning Electron Microscopy-Energy Dispersive Spectroscopy
(SEM-EDS) mapping results for alumina highlighted in pink (left) and zirconium
highlighted in green (right) on the 3.1%ZrO2-Al2O3 support.Figure 4.9
Figure 4.9: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results for 3.1%ZrO2-Al2O3 where Al2O3 is pink (left) and ZrO2 is
green (right).
Figure 4.10: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results overlaid on scanning electron microscopy image for
3.1%ZrO2-Al2O3 where Al2O3 is pink (left) and ZrO2 is green (right).
NASA/TM—2012-216020 42
Figure 4.11: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) mapping results for 3.1%ZrO2-Al2O3 overlaid on scanning electron microscopy
image where Al2O3 is pink (left) and ZrO2 is green (right).
NASA/TM—2012-216020 43
Figure 4.12: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) quantitative results for 3.1%ZrO2-Al2O3
Table 4.3: Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDS) quantitative results for 3.1%ZrO2-Al2O3.
Element Wt% At% K-Ratio Z A F
AlK 91.23 97.23 0.8451 1.0104 0.9157 1.0013
ZrL 8.77 2.77 0.0336 0.8724 0.4387 1.0000
NASA/TM—2012-216020 44
4.2 Brunauer Emmet Teller (BET)4 and Barrett Joyner Halenda (BJH)33
Measurements
Surface area measurement by nitrogen adsorption and desorption results are
shown in Table 4.4. The surface areas and average adsorption pore radii of Puralox
TH 100/150 Ti10, Puralox SCFa-200 Zr3, and Puralox SCFa-140/L3 supports were
measured to be 137.94 m2/g and 13.327 nm, 152.63 m2/g and 3.74 nm, and 142.26
m2/g and 6.27 nm respectively. The baseline support surface areas can be used to
calculated the expected surface area of each support after the cobalt was loaded.
Since BET was completed on the catalysts before reduction, the surface areas take
into account cobalt oxide not cobalt metal. In order to determine potential pore
blockage on the surface, the catalyst cobalt weight percents must be corrected to
cobalt oxide weight percents. A weight percent of 15% cobalt metal is equivalent to
20% cobalt oxide (Co3O4). If we assume that the support is the only contributor to
area, then the area of 15%Co/9.7%TiO2-Al2O3 would be expected to be 0.80 x
137.94m2/g (the area of 9.7%TiO2-Al2O3)= 110.35 m2/g, which is within
experimental error of the measured surface area of 113.26 m2/g. Similarly, the area
of 15%Co/3.0%La2O3-Al2O3 should be 0.8 x 142.26 m2/g (the surface area of
3.0%La2O3-Al2O3) = 113.81 m2/g, which also corresponds to the measured surface
area of 124.41 m2/g quite well. Likewise, the area of 15%Co/3.1%ZrO2-Al2O3
should be 0.8 x 183.5 m2/g (the area of 3.1%ZrO2-Al2O3)=146.8 m2/g, which
matches the measured surface area of 152.63 m2/g reasonably well. These results
can be compared more clearly in Figure 4.13.
NASA/TM—2012-216020 45
Figure 4.13: Expected BET surface areas of (a) 15%Co/9.7%TiO2-Al2O3, (b)15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-Al2O3
The BET surface area results of the supports and their corresponding
catalysts are shown in Figure 4.14. The surface area of the catalyst in relation to it’s
bare support was expected to be lower, which was shown to be true for all supports
upon addition of cobalt.
0
20
40
60
80
100
120
140
160
180
a b c
Surf
ace
Are
a (m
m2/
g)
expected actual
NASA/TM—2012-216020 46
Figure 4.14: BET surface area comparison of (a) 15%Co/9.7%TiO2-Al2O3 , (b)15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-Al2O3, (d) 15%Co/Al2O3
HP14/150, (e) 15%Co/Al2O3 SBA150, and (f) 15%Co/Al2O3 SBA200.
Along with surface area measurements, Table 4.4 also displays the BJH
adsorption average pore radius. The BJH adsorption average pore radius data is
also displayed in Figure 4.15 and the BJH desorption average pore volume is shown
in Figure 4.16. In comparing each adsorption average pore radius, the cobalt
loading decreased the pore radius on all three supports. The most significant
reduction in pore radius was seen in the loading of 15% cobalt on the 9.7%TiO2-
Al2O3 support, where the support and catalyst pore radii were measured to be 13.37
nm and 10.92 nm respectively, which also had the largest pore radius. Since the
pore radius decreased upon addition of cobalt to all of the supports, the data
suggests that the pores were filled uniformly. The pore volume decreased slightly
upon addition of cobalt to the support, as expected.
0
50
100
150
200
250
a b c d e f
Surf
ace
Are
a (m
m2/
g)
support catalyst
NASA/TM—2012-216020 47
Figure 4.15: Barrett Joyner Halenda (BJH) average pore radius adsorption data for(a) 15%Co/9.7%TiO2-Al2O3 , (b) 15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-
Al2O3, (d) 15%Co/Al2O3 HP14/150, (e) 15%Co/Al2O3 SBA150, and (f) 15%Co/Al2O3SBA200.
Figure 4.16: Barrett Joyner Halenda (BJH) average pore volume desorption data for(a) 15%Co/9.7%TiO2-Al2O3 , (b) 15%Co/3.0%La2O3-Al2O3, (c) 15%Co/3.1%ZrO2-
Al2O3, (d) 15%Co/Al2O3 HP14/150, (e) 15%Co/Al2O3 SBA150, and (f) 15%Co/Al2O3SBA200.
0
2
4
6
8
10
12
14
16
a b c d e f
Ave
rage
Por
e Ra
dius
(nm
)
support catalyst
0
0.2
0.4
0.6
0.8
1
1.2
a b c d e f
Des
orpt
ion
Ave
rage
Por
eVo
lum
e (c
m3/
g)
support catalyst
NASA/TM—2012-216020 48
Table 4.4: BET surface area measurements and BJH pore volume and pore radiusmeasurements
Figure 4.26: Pulse reoxidation cluster size data, which is corrected for extent ofreduction for each catalyst in comparison to it’s reference catalysts.
(7) Jacobs, G.; Das, T. K.; Zhang, Y.; Li, J.; Racoillet, G.; Davis, B. H. AppliedCatalysis A: General 2002, 233, 263.
(8) Ma, W.; Jacobs, G.; Sparks, D. E.; Gnanamani, M. K.; Pendyala, V. R. R.; Yen, C.H.; Klettlinger, J. L. S.; Tomsik, T. M.; Davis, B. H. Fuel 2011, 90, 756.
(9) Khodakov, A. Y.; Lynch, J.; Bazin, D.; Rebours, B.; Zanier, N.; Moisson, B.;Chaumette, P. Journal of Catalysis 1997, 168, 16.
(10) Iglesia, E.; Soled, S. L.; Fiato, R. A.; Via, G. H. Journal of Catalysis 1993, 143,345.
(11) Espinoza RL, V. J., van Berge PJ, Bolder FH; Sastech (Proprietary) Limited:Johannesburg, South Africa, 1998.
(12) Wang, W. J.; Chen, Y. W. Applied Catalysis 1991, 77, 223.
(13) Murzin, D. Y. Chemical Engineering Science 2009, 64, 1046.
(14) Jacobs, G.; Ribeiro, M. C.; Ma, W.; Ji, Y.; Khalid, S.; Sumodjo, P. T. A.; Davis, B. H.Applied Catalysis A: General 2009, 361, 137.
(15) Khodakov, A. Y.; Chu, W.; Fongarland, P. Chemical Reviews 2007, 107, 1692.
(25) Rohr, F.; Lindvåg, O. A.; Holmen, A.; Blekkan, E. A. Catalysis Today 2000, 58,247.
(26) Feller, A.; Claeys, M.; Steen, E. v. Journal of Catalysis 1999, 185, 120.
(27) 2009, 52, 1044.
(28) Ulín, C. A.; De Los Reyes, J. A.; Escobar, J.; Barrera, M. C.; Cortés-Jacome, M. A.Journal of Physics and Chemistry of Solids 2010, 71, 1004.
(29) Xiong, H.; Zhang, Y.; Liew, K.; Li, J. Journal of Molecular Catalysis A: Chemical2005, 231, 145.
(30) Niemantsverdriet, J. W. Spectroscopy in Catalysis; Third ed.; WILEY-VCH,2007.
(31) Feltes, T. E.; Espinosa-Alonso, L.; Smit, E. d.; D'Souza, L.; Meyer, R. J.;Weckhuysen, B. M.; Regalbuto, J. R. Journal of Catalysis 2010, 270, 95.
(32) Tao, C.; Li, J.; Zhang, Y.; Liew, K. Y. Journal of Molecular Catalysis A: Chemical2010, 331, 50.
NASA/TM—2012-216020 78
(33) Barrett, E. P.; Joyner, L. G.; Halenda, P. P. Journal of the American ChemicalSociety 1951, 73, 373.
(34) Jacobs, G.; Ji, Y.; Davis, B. H.; Cronauer, D.; Kropf, A. J.; Marshall, C. L. AppliedCatalysis A: General 2007, 333, 177.
(35) Simionato, M.; Assaf, E. M. Materials Research 2003, 6, 535.
(36) van Steen, E.; Claeys, M.; Dry, M. E.; van de Loosdrecht, J.; Viljoen, E. L.;Visagie, J. L. The Journal of Physical Chemistry B 2005, 109, 3575.
(37) Jacobs, G.; Patterson, P. M.; Das, T. K.; Luo, M.; Davis, B. H. Applied Catalysis A:General 2004, 270, 65.
NASA/TM—2012-216020 79
APPENDIX A. Pore Size Distribution Profiles
Figure 4.34: Adsorption pore size distribution of (left) 9.7%TiO2-Al2O3 and (right)15%Co/9.7%TiO2-Al2O3.
Figure 4.35: Desorption pore size distributions of 9.7%TiO2-Al2O3(left) and15%Co/9.7%TiO2-Al2O3(right)
Figure 4.36: Adsorption pore size distribution of 3.0%La2O3-Al2O3 (left) and15%Co/3.0%La2O3-Al2O3 (right)
NASA/TM—2012-216020 80
Figure 4.37: Desorption pore size distribution of 3.0%La2O3-Al2O3 (left) and15%Co/3.0%La2O3-Al2O3 (right)
Figure 4.38: Adsorption pore size distribution of 3.1%ZrO2-Al2O3 (left) and15%Co/3.1%ZrO2-Al2O3 (right)
Figure 4.39: Desorption pore size distributions of 3.1%ZrO2-Al2O3(left) and15%Co/3.1%ZrO2-Al2O3 (right).
NASA/TM—2012-216020 81
(a) (b)
(c) (d)
Figure 4.40: Adsorption pore size distribution of Al2O3 HP14/150 (a)& (b) and15%Co/Al2O3 HP14/150 (c) & (d).
NASA/TM—2012-216020 82
(a) (b)
(c) (d)
Figure 4.41: Desorption pore size distribution of Al2O3 HP14/150 (a)& (b) and15%Co/Al2O3 HP14/150 (c) & (d).
NASA/TM—2012-216020 83
(a) (b)
(c) (d)
Figure 4.42: Adsorption pore size distributions of (a) & (b): SBA 150 Al2O3 and (c) &(d): 15%Co/SBA 150 Al2O3.
NASA/TM—2012-216020 84
(a) (b)
(c) (d)
Figure 4.43: Desorption pore size distributions of (a) & (b): SBA 150 Al2O3 and (c) &(d): 15%Co/SBA 150 Al2O3.
NASA/TM—2012-216020 85
(a) (b)
(c) (d)
Figure 4.44: Adsorption pore size distributions of (a) & (b): Al2O3.SBA 200 and (c) &(d): 15%Co/Al2O3.SBA 200
NASA/TM—2012-216020 86
(a) (b)
(c) (d)
Figure 4.45:Desorption pore size distributions of (a) & (b): Al2O3.SBA 200 and (c) &(d): 15%Co/Al2O3.SBA 200